Post on 19-Jul-2020
AUTHORS
Richard W. Hurst � Hurst & Associates,Inc., 9 Faculty Court, Thousand Oaks, Califor-nia 91360; Alasrwh@aol.com
Richard W. Hurst received his doctorate in ge-ology and geochemistry from the Universityof California, Los Angeles in 1975. He is a pro-fessor of geological sciences and geochemistryat California State University, Los Angeles. Since1978, his interests have focused on perform-ing research and consulting in the field offorensic environmental geochemistry, apply-ing isotope geochemistry, hydrogeologic data,and statistical methods to resolve problemsasssociated with environmental remediationand petroleum exploration.
Gene W. Schmidt � Gene W. Schmidt Envi-ronmental Consulting, 11619 S. Hudson Place,Tulsa, Oklahoma 74137; envirodog@aol.com
Gene W. Schmidt retired as director of Ground-water Management Services in Amoco Corpora-tion’s Environment, Health, and Safety Departmentin 1992. He currently operates an enviromentalconsulting firm that specializes in the forensicsof petroleum hydrocarbon contamination. Heholds two undergraduate degrees (geology andanalytical chemistry) and two graduate degrees(organic and aqueous geochemistry and geo-chemistry). Prior to heading the formation waterprogam at Amoco, he conducted groundwaterpollution control work for the Kansas State Boardof Health. He is a certified groundwater profes-sional in the Association of Groundwater Scientistsand Engineers, a professional hydrogeologistin the American Institute of Hydrology, and aregistered professional geologist.
ACKNOWLEDGEMENTS
We thank J. Berton Fisher for his initial sugges-tion and encouragement to publish our re-sults from site-specific investigations andIan Kaplan for discussions and debate withregard to estimating ages of environmentalreleases of petroleum hydrocarbons. We areindebted to Bruce Torkelson for the high-resolution gas chromatograms needed for thisresearch and A. Jerome Skarnulis of ComputerDesign Software, Inc., for donating the timenecessary to perform the transformations re-quired to more accurately integrate the originalChristensen and Larsen (1993) data with ours.
Age significance of nC17/Prratios in forensic investigationsof refined product and crudeoil releasesRichard W. Hurst and Gene W. Schmidt
ABSTRACT
Since the 1980s, several techniques have been developed to esti-
mate the year a petroleum hydrocarbon release occurred. In this
article, we evaluate and expand on the model of Christensen and
Larsen, who proposed that the degradation of normal heptadecane
relative to pristane (i.e., nC17/Pr ratios) could be used to estimate
the age of diesel fuels released into the environment. Linear re-
gression analyses of nC17/Pr ratios from known subsurface releases
of crude oil, middle distillate, fuel oil, and lubricating oil in diverse
climatic settings (Subarctic, temperate, and tropical) define a statis-
tically significant, negative linear correlation termed the middle dis-
tillate degradation (MDD) model, in which, like the Christensen-
Larsen model, nC17 is almost entirely degraded within about 20 yr.
By comparison, our investigations indicate that degradation of nC17
relative to Pr in aerobic, surface environments is also systematic,
following a first-order kinetic relationship in which nC17 degrades
about 5–6 yr subsequent to the release. As observed by others, the
timescale of degradation under aerobic conditions is accelerated.
We also present analyses of average initial (nC17/Pr)o ratios of
about 4500 worldwide crude oils and 90 domestic refined products
(diesel jet A, fuel oils) to evaluate how variations in this parameter
impact MDD model ages. As stipulated in debates surrounding the
original Christensen-Larsen model, applications of the MDD model
should be evaluated carefully on a case-by-case basis and not in an ad
hoc fashion. Our results not only provide a database for evaluating
the significance of geographic variations in (nC17/Pr)o ratios but also
allow experienced investigators to estimate MDD model age uncer-
tainties (3–10-yr window of uncertainty under optimal to worst
case conditions, respectively) at sites where it is determined that
such models are applicable.
Environmental Geosciences, v. 12, no. 3 (September 2005), pp. 177–192 177
Copyright #2005. The American Association of Petroleum Geologists/Division of EnvironmentalGeosciences. All rights reserved.
DOI:10.1306/eg.04260404004
INTRODUCTION
In 1993, Christensen and Larsen published a model
(hereafter, the Christensen-Larsen [C-L] model), which
they proposed could estimate the age of diesel releases
to about ±2 yr. The C-L model was calibrated using
measured ratios of n-heptadecane to 2,6,10,14-tetra-
methylpentadecane (i.e., n-C17 and pristane, Pr, respec-
tively) extracted from soils impacted by diesel releases
of known age (�5–20 yr old) in Denmark. Christensen
and Larsen observed a linear relationship between the
nC17/Pr ratio of weathered diesel and time elapsed since
the release. The relationship, as approximated by Ka-
plan et al. (1997), is
T ðyearsÞ ¼ �8:4ðnC17=PrÞ þ 19:8
The observed decrease in the nC17/Pr ratio over time
was attributed to the preferential degradation of the
n-alkane, nC17, relative to pristane, an isoprenoid hy-
drocarbon. Christensen and Larsen (1993) also analyzed
11 fresh diesel samples to assess variations in initial
nC17/Pr ratios (hereafter (nC17/Pr)o); they observed a
range of 1.5–2.5 (95% confidence level standard error
of the mean) and an average value of 1.98.
Given the importance of accuracy and scientific
defensibility of models designed to estimate the age of
petroleum hydrocarbon releases, it is not surprising
that the applicability and scientific validity of the C-L
model have been the subject of debate (Kaplan and
Galperin, 1996; Schmidt, 1996, 1998; Kaplan et al., 1997;
Morrison, 2000a, b; Smith et al., 2001; Wade, 2001;
Alimi, 2002; Stout et al., 2002; Wade, 2002; Kaplan,
2003). These authors note that variations in (nC17/Pr)o
of diesel fuel released at a site and the rate of degra-
dation of nC17 relative to Pr, the latter a function of
local site conditions (Bossert and Bartha, 1984; Alvarez
and Vogel, 1995; Alvarez et al., 1998), contribute to un-
certainties in C-L model age estimates. Wade (2001,
2002) observed that investigations concerning the ap-
plicability of the C-L model to releases at sites where
conditions differ significantly from those in Denmark
have not been performed. The two extreme views re-
garding the applicability of the C-L model range from
those of Smith et al. (2001), who conclude the C-L
model is totally inappropriate in any situation, to those
of Schmidt (1996, 1998), Wade (2001, 2002), and Ali-
mi (2002), who suggest that the C-L model may be
used, with caution, to estimate the age of middle dis-
tillate, heavy fuel oil, and crude oil releases in some
areas of the United States and Europe.
In this article, we will address these uncertain-
ties, providing data on variations in (nC17/Pr)o and in-
corporating data from sites whose climatic conditions
range from temperate to tropical to assess the impact
on the rate of degradation and resultant C-L model
age estimates. Measured (nC17/Pr)o ratios of about
4500 worldwide crude oils and 90 domestic petroleum
products are presented to provide constraints on the
statistical variation in this parameter and its impact
on resultant model ages. The original C-L model data
(Christensen and Larsen, 1993) have been reproduced
by transformation (scanning and scaling into a coordi-
nate system; A. J. Skarnulis, 2003, personal communi-
cation), so that coordinates of each datum point could
be determined. Their data have been integrated with
data from Schmidt (1996, 1998), Wade (2001), Hurst
(this study), and new data from documented middle
distillate and crude oil releases from diverse geographic
regions in the United States and the Caribbean to assess
potential effects of subsurface temperature variations
and spill location, i.e., subsurface vs. surface, on the deg-
radation of nC17 relative to pristane.
The culmination of these data is a modified C-L
model, the middle distillate degradation (MDD) mod-
el, for consideration as a tool to evaluate the age of
middle distillate, heavy fuel oil, and crude oil releases
up to about 20 yr old. The MDD model should be
applied on a case-by-case basis, heeding the concerns
noted (references above) regarding ad hoc applications
of any model in environmental forensic investigations.
A second article, Schmidt and Hurst (unpublished data)
will focus on case studies in which the MDD model
was used either independently or in conjunction with
the anthropogenic lead archeostratigraphy model of
Hurst (2000, 2002, 2003) to estimate ages of middle
distillate and contemporaneous leaded aviation gaso-
line releases at sites in the United States.
HIGH-RESOLUTION GAS CHROMATOGRAPHY
Accurate and precise measurements of nC17/Pr ratios
require high-resolution gas chromatography (HRGC),
in which individual peaks are identified and resolved,
peak overlap and asymmetry are minimized, and signal-
to-noise ratios are high. It is also imperative that
standards are analyzed routinely to calibrate the gas
chromatograph, assess baseline drift, and evaluate the
reproducibility of the instrument (Dyson, 1995). For
reference, representative HRGCs of fresh diesel and jet
A are shown in Figure 1a and b, respectively; these can
178 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
be compared to chromatograms of weathered diesel
and jet A (Figure 2a, b, respectively). Insets depict peak
shape and baseline (dashed line) details at the reten-
tion times where nC17 and Pr elute from the column.
The weathered products were sampled from releases
that are in excess of 15 yr old.
As observed in the HRGC insets of both the fresh
and weathered products, peak intensities of the noise
(i.e., small peaks above the baseline in the valleys be-
tween the nC17-Pr peaks) never exceed about 2 mV;
hence, for fresh to moderately weathered free product,
the signal-to-noise ratio is high; this ratio decreases as
weathering proceeds. The nC17 and Pr peaks are well
resolved, with negligible tailing and asymmetry even
in the case of weathered product, where the intensity
of the unresolved complex mixture has increased sig-
nificantly relative to the peak heights of paraffins and
isoprenoids (Figure 2a, b). When the nC17/Pr ratio is
approximately 0.1, resulting corrections to the nC17
peak caused by peak overlap produce a higher uncer-
tainty in the nC17/Pr ratio.
Measured (nC17/Pr)o ratios of the fresh diesel and
jet A (2.0 and 2.3; Figure 1a, b, respectively) are com-
parable to those measured on 92 domestic middle dis-
tillates and heavy fuel oils (2.0 ± 0.2; discussed later in
this study). Peak intensities, and therefore concentra-
tions, of nC17 and Pr are lower in jet A than in diesel
(sample volumes injected for analysis are identical);
however (nC17/Pr)o ratios of jet A fuels analyzed in
this study, like diesel, average approximately 2.0, re-
flecting those of their source crude oil (Schmidt, 1996,
1998; Morrison, 2000c; this study). Furthermore, al-
though the boiling points of nC17 and Pr (302 and
296jC respectively [�570jF]; http://physchem.ox.ac
.uk/MSDS) are similar to the cutoff temperature of
jet A (�300jC; Kaplan, 2003), we have not observed
any evidence for preferential fractionation of nC17
from Pr during refining of jet A that would result in
higher variability in jet A (nC17/Pr)o ratios relative to
that observed in diesel fuels. Continued distillation of
crude oil to about 400jC (�750jF), the cutoff tem-
perature for diesel fuel, produces the higher concen-
trations of nC17 and Pr present in diesel fuels.
DEVELOPMENT OF THE MDD MODEL
The new nC17/Pr ratios presented in this study were
analyzed in soil and weathered product; ages of the
hydrocarbon releases ranged from about 5 to 20 yr old.
In the case of soils, samples were collected at depths
exceeding 4–5 m (13–16 ft), and residual total petro-
leum hydrocarbon concentrations were about 125–
800 mg/kg (ppm).
New Site Details
Summaries of specific details regarding the type of hy-
drocarbon release, year of the release, year(s) of sam-
pling, nature of the matrix analyzed, and average (nC17/
Pr) ratios measured on five to seven samples are pre-
sented below. The range in measured (nC17/Pr) ratios,
as determined from the standard error of the mean at
the 95% confidence level, is shown in parentheses next
to each average value. These data are plotted in Figure 3
as either the continental United States or the Caribbean
data points.
West Texas
Inventory records indicated that a supply line to a pro-
duction pipeline had released crude oil at two different
locations and on two separate occasions; distances be-
tween the locations of the releases and site remediation
data indicated the presence of two distinct plumes. The
older release dated back to late 1975, whereas the youn-
ger release occurred in 1990. Sampling of free product
and analysis by gas chromatography were performed
in 1994 to evaluate the extent of the contamination.
Measured (nC17/Pr) ratios of degraded crude oil from
the 1975 release averaged 0.15 (0.08–0.19), whereas
those associated with the 1990 release averaged 1.89
(1.85–1.91).
In a separate incident, a refueling station pipeline
in El Paso, Texas, released diesel fuel into the subsur-
face in 1990. Soil samples collected in 1994 yielded an
average (nC17/Pr) ratio of 1.8 (1.7–1.87).
Maryland and Georgia
Diesel fuel releases were documented at a Maryland
fuel terminal (1978) and from a pipeline in Geor-
gia (1986). Samples of hydrocarbon-impacted soils
were collected in 1993 and 1994, respectively. Aver-
age (nC17/Pr) ratios from Maryland samples averaged
0.53 (0.35–0.60); samples in Georgia averaged 1.34
(1.25–1.41).
Pennsylvania
A machine shop had a release of lubricating oil in
1978 that permeated the porous soils that comprised
the floor. The business was closed, but the structure
Hurst and Schmidt 179
remained; no efforts were made at that time to reme-
diate the release. In 1993, sampling of soils indicated the
presence of lubricating oil whose measured (nC17/Pr)
ratios averaged 0.28 (0.24–0.35).
Caribbean
Diesel releases from damaged tanks and pipelines oc-
curred as a result of severe hurricanes between 1982
and 1983. A part of the free product from disrupted
subsurface pipelines migrated rapidly into the subsur-
face, impacting groundwater in the region. Free prod-
uct samples were collected in 1997 and 2002 from
the same monitoring wells to evaluate the extent of
degradation of the diesel in the subsurface. Measured
(nC17/Pr) ratios of product collected in 1997 aver-
aged 0.65 (0.44–0.72); results from the product in
the same monitoring wells taken in 2002 average 0.13
(0.05–0.22).
Correlation of nC17/Pr with the Age of the Release:Earlier Investigations
To evaluate the geographic applicability of the C-L mod-
el, comparative linear regression analyses of nC17/Pr
vs. T, the known age of a middle distillate release, were
performed on each data set from the original Chris-
tensen and Larsen (1993) article, the work of Schmidt
(1996, 1998), and recent results from Wade (2001).
Each datum point plotted in Christensen and Larsen
(1993) has been quantified through transformation to
acquire specific coordinates, nC17/Pr and release age,
for each datum point. Linear regression analyses were
performed using ISOPLOT, a program developed
by the U.S. Geological Survey (Brooks et al., 1972;
Ludwig, 1983) that accounts for analytical error in
each datum point to produce the best-fit line; out-
put from ISOPLOT includes the 95% confidence level
Figure 1. Representative high-resolution gas chromatograms of fresh (a) diesel and (b) jet A fuels. Note the excellent separation ofnC17 from pristane (inset) and the high signal-to-noise ratio. Baseline is shown as a dashed line.
180 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
errors in the slope and intercept as shown in Table 1
and plotted in Figure 3. The value of R2, as deter-
mined from standard linear regression programs, is
also provided.
Linear regression analyses of each independent data
set from Christensen and Larsen (1993), Schmidt (1996,
1998), and Wade (2001) produce significant negative
correlations between nC17/Pr ratios and T (R2 = 0.896–
0.979). Although resultant slopes and intercepts dif-
fer, they are statistically indistinguishable when 95%
confidence level errors are considered. Maximum model
age variations, as determined by inputting a specific
(nC17/Pr) ratio into the C-L, Schmidt, and Wade mod-
els (Table 1; previous investigations), range from 0.7
to 1.7 yr. The largest discrepancy is observed when
(nC17/Pr) equals 2; the Wade model yields an age of
�0.1 yr compared to the C-L model age of 1.6 yr; given
the uncertainties in these models (±1.4 to ±1.87 yr,
respectively; Table 1), the 1.7-yr difference in model
ages is not significant. Theoretically, as nC17/Pr ap-
proaches zero, variations among the model ages range
from 0.3 to 0.9 yr, determined by differences in values
of To. These variations, once again, lie well within the
reported age uncertainty associated with the model,
±2 yr, as originally proposed by Christensen and Larsen
(1993) and corroborated by both Schmidt (1996, 1998)
and Wade (2001).
It is significant that the slopes of the linear re-
gressions, i.e., degradation rate of nC17 relative to Pr,
from the independent investigations (Table 1), are
statistically indistinguishable, given the different geo-
graphic locations of each study. Wade (2001) evaluated
the C-L model, as originally published, at sites in the
northeastern United States, New England (Connecticut,
Massachusetts, Maine, New Hampshire, Rhode Island,
and Vermont), where surface temperature ranges and
precipitation are generally comparable to those of Den-
mark. Schmidt (1996, 1998) evaluated the C-L model
Figure 1. Continued.
Hurst and Schmidt 181
using domestic crude oil and diesel releases in areas
where climate differs significantly from that of Denmark
and the northeastern United States. Given good agree-
ment between model ages determined by the C-L model
and documented years of diesel and crude oil spills, both
Wade and Schmidt concluded that the C-L model could
potentially be applied to estimate ages of petroleum
hydrocarbon releases in the United States.
Unified (MDD) Model: Subsurface Releases
Given the observed statistical concordance among the
results of the linear regression analyses from the three
independent studies, the three data sets and results
from the new site analyses were combined and then
regressed using ISOPLOT to generate the unified,
MDD model (Table 1, Unified Model; Figure 3). Be-
cause the number of data points regressed has increased
but the magnitude of scatter has not, the 95% confi-
dence level errors in the MDD model slope and inter-
cept decrease.
As observed in Table 1, 95% confidence level un-
certainties associated with the MDD model linear re-
gression are improved, approximately by factor of 2,
relative to those of previous investigations (Christensen
and Larsen, 1993; Schmidt, 1996, 1998; Wade, 2001).
Absolute age variations between MDD model ages vs.
those determined using the Christensen-Larsen, Schmidt,
or Wade regressions, as calculated for nC17/Pr ratios
of 2.0 and 0.2, range from 1.2 to 0.4 yr, respectively;
all models yield identical results, 11 yr, when nC17/Pr
equals 1.0.
The MDD model age resolution could theoreti-
cally approach ±1 yr if all of the following conditions
Figure 2. Representative high-resolution gas chromatograms of weathered (a) diesel and (b) jet A fuels. Although nC17/Pr ratios areless than 0.4 in these products, nC17 and pristane are readily resolved from each other and the baseline (dashed line); note that peakoverlap is now apparent because of the unidentified peak located between nC17 and pristane.
182 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
were satisfied: (1) (nC17/Pr)o ratios could be determined
with reasonable certainty; (2) a sufficient number of
statistically representative samples were analyzed; and
(3) information pertinent to local environmental con-
ditions were available. Being conservative, we recom-
mend that the uncertainty be increased to ±1.5 yr, a
3-yr window, under optimal conditions. This increase
also improves the statistical fit of the calibration data
points to the MDD model regression shown in Figure 3;
31 of 35 (89%) calibration data points lie within the
± 1.5-yr confidence level interval as compared to 25
of 35 (71%), when the confidence level interval is
±1 yr. The effect of uncertainties in (nC17Pr)o ratios
on MDD model age resolution is discussed in the sub-
sequent section.
The development of the MDD model implements
the suggestion to evaluate the degradation of hydro-
carbon releases in geographic regions where subsur-
face temperatures differ from those of Denmark
(Wade, 2001, 2002); our work expands potential ap-
plications to estimating release ages for lubricating and
crude oils, which have not been investigated (Kaplan,
2003). Because the degradation rate of nC17 relative
to pristane observed in crude oil and diesel releases
from Georgia, Texas, and the Caribbean is statisti-
cally indistinguishable from those determined in Den-
mark and the northeastern United States (Figure 3)
(Christensen and Larsen, 1993; Schmidt, 1996, 1998;
Wade, 2001; this study), we conclude that the ob-
served temperature range in shallow soils and ground-
water in a large part of the United States, about
8 to 20jC (Collins, 1925; Schmidt and Marsi, 1961;
Heath, 1984; Hern and Melancon, 1986; van der Lee-
den et al., 1990), does not necessarily alter the degra-
dation rate of nC17 relative to pristane in subsurface
environments.
Figure 2. Continued.
Hurst and Schmidt 183
VARIATIONS IN (nC17/Pr)o RATIOS ANDUNCERTAINTY IN MDD MODEL AGES
In this section, the uncertainties in MDD model ages
that are dependent on the extent of our knowledge of
(nC17/Pr)o ratios at a particular site will be addressed
through a series of scenarios. Conditions will range from
those whose optimal (nC17/Pr)o ratios are known or can
be adjusted to account for variations in (nC17/Pr)o to
those where a broad range in (nC17/Pr)o ratios, from
about 1.6 to 2.6, might be used to estimate (nC17/Pr)o
in cases where sampling of the suspected source of a
release is no longer viable.
Figure 3. Middle distillate degradation model: nC17/Pr vs. known ages of subsurface and surface releases of crude oil, diesel, jet A,and fuel oils. Data plotted are derived in part from the published literature (Denmark: Christensen and Larsen, 1993; northeasternUnited States: Wade, 2001; surface (AK/PA): Hostettler and Kvenvolden, 1994; Smith et al., 2001) as well as this work (all geographiclocations except Denmark). The decrease in nC17/Pr ratios as a function of time in crude oil and refined product releases in thesubsurface define a well-correlated, linear trend (MDD model); the dotted error envelope around the best-fit linear regression indicatesthe 97.5% confidence level uncertainty, i.e., ±1.5 yr. Surface releases of crude oil and middle distillate fuels (open symbols) define afirst-order degradation curve in which nC17 is almost entirely degraded within approximately 5–6 yr.
Table 1. Comparative Linear Regression Analyses (nC17/Pr vs. Time)
T (years) = �l (nC17/Pr) + To
Name of Model Number of Data Points l (Slope)* To R 2 (nC17/Pr)o**
I. Previous investigationsChristensen and Larsen (1993) 13 9.47 ± 1.87 20.5 ± 2.1 0.896 2.16 ± 0.22
Schmidt (1996, 1998) 9 9.36 ± 1.53 20.2 ± 2.4 0.967 2.16 ± 0.25
Wade (2001) 9 10.6 ± 1.40 21.1 ± 1.4 0.979 1.99 ± 0.13
II. Unified modelMDD (this study) 35 9.76 ± 0.81 20.7 ± 1.0 0.952 2.12 ± 0.12
*Data have been fitted using ISOPLOT, a linear regression program that considers analytical errors associated with each datum point in the statistical fitting of the best-fit line (Brooks et al., 1972; Ludwig, 1983). The program calculates 95% confidence level errors in the resultant slope and intercept as shown.
**The (nC17/Pr)o ratio is calculated from each linear regression model (i.e., the X-intercept when nC17/Pr = 0).
184 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
Referring to Table 1, the MDD model regression
(Table 1) is given by
T ¼ �9:76 ðnC17=PrÞ þ 20:7
Setting T equal to zero, we obtain the following re-
lationship between To and (nC17/Pr)o:
To ¼ 9:76 ðnC17=PrÞo
Based on the results of this study, the rate of degrada-
tion of nC17 relative to Pr in subsurface environments,
i.e., the slope of the MDD model, appears to be con-
stant. Thus, to adjust the MDD model regression for
variations in (nC17/Pr)o while maintaining the slope at
9.76, we use the relationship between To and (nC17/
Pr)o to calculate a new value for To. The value of To
sets the upper limit to the oldest release that can be
dated for a given value of (nC17/Pr)o. For example, if
(nC17/Pr)o of a release is known to be 1.5, To will equal
14.6 yr; for an (nC17/Pr)o ratio of 4, To equals 39 yr. The
latter example is interesting, in that it suggests that the
age of hydrocarbon releases greater than 20 yr old
might be modeled in cases where (nC17/Pr)o of a release
exceeds that of the MDD model (i.e., 2.12; Table 1). The
data presented in this study do not allow us to evalu-
ate dating releases greater than 20 yr old; in scenario 1
below, however, we provide an example using data
from the northwestern United States plume discussed
earlier, in which the MDD model age has been ad-
justed because the (nC17/Pr)o of the diesel release ex-
ceeds that of the MDD model.
The scenarios are provided as suggested guidelines,
providing limitations on the age resolution achievable
using the MDD model given the variations in (nC17/Pr)o.
Each scenario assumes that a statistically representa-
tive number of samples were analyzed, that sampling
protocols are followed, and that site-specific environ-
mental conditions have been evaluated.
Scenario 1: (nC17/P)o is Known
If the (nC17/Pr)o ratio of a hydrocarbon released at a
site is known to be equivalent to that of the MDD
model, 2.1 ± 0.1, the age uncertainty could theoret-
ically approach ±1 yr. As stated earlier, being conser-
vative, an uncertainty of ±1.5 yr, i.e., 3-yr window,
would be reasonable under optimal conditions; no ad-
justment to resultant MDD model ages, as determined
using the regression in Table 1, is required. As stated
above, what if (nC17/P)o is known, but it is not equal to
that of the MDD model? The northwestern United
States plume discussed earlier provides an example of
such a situation.
Case Study: Northwestern United States
Limitations on oxygen, water, and nutrients in the in-
terior of a hydrocarbon plume have been suggested
as inhibiting the rate at which hydrocarbons, including
nC17 and Pr, degrade, thereby allowing hydrocarbons
to persist in the environment for decades (Bossert and
Bartha, 1984); we refer to these types of environments
as conservative environments. Although such conser-
vative environments may occur, the volume of lit-
erature published on hydrocarbon degradation indi-
cates that such environments are not the norm. To
address this issue in more detail, we investigated a
large gasoline-diesel plume in the northwestern United
States to (1) evaluate the effect of location within a
free product plume on degradation rates; (2) study de-
gradation of free product over a 13-yr interval; and
(3) assess the impact of a significantly different (nC17/
Pr)o ratio on model ages (discussed later in this study).
The initial release of gasoline and diesel fuels is
suspected to have started in the early 1970s as the result
of a transfer line leak; the release continued until its
detection in 1983 when remediation commenced. The
dimensions of the about 10–15-yr-old release were
1350 m (4429 ft) in length, 125–400 m (400–1300 ft)
in width, and up to 1.5 m (4.5 ft) in thickness. Given the
plume length, 1350 m (4429 ft), and average ground-
water velocity, about 0.45 m/day (1.5 ft/day), ground-
water requires approximately 8 yr to migrate from the
source of the release to the plume terminus. Although
such inverse modeling is sometimes used to estimate
the age of a release, contaminant plumes do not advect
with groundwater, hence, inverse modeling commonly
underestimates the age of a release (Morrison, 2000b;
this study) as observed in this case.
A series of 12 free product samples were collected
from groundwater monitoring wells at the following
distances downgradient from the source: 250, 675, 1045,
and 1350 m (820, 2214, 3428, and 4429 ft, respec-
tively). Gas-chromatographic analyses were performed
on each free product sample. The resulting nC17/Pr ra-
tios (samples from plume margins are italicized) and
downgradient distance for each of the 12 samples are
shown in Table 2.
The steady decrease in diesel-free product nC17/Pr
ratios correlates well with the distance from the source
Hurst and Schmidt 185
(R2 = 0.998) and, therefore, the time required for the
free product to migrate from the point of the release.
As observed, nC17/Pr ratios of the free product
sampled from the interior of the plume 250 and 675 m
(820 and 2214 ft) downgradient from the source are
slightly higher than those at the plume margins; how-
ever, nC17/Pr ratios of the free product from the plume
interior vs. plume margins are equivalent in samples
collected 1350 m (4429 ft) downgradient. The nC17/Pr
ratios of free product 1045 m (3428 ft) downgradient
were all collected at the plume margin, yet the degra-
dation of nC17 relative to Pr at this distance from the
source remains consistent with that of the other sam-
ples. Variations in nC17/Pr ratios of the product sam-
pled at the plume margin vs. the interior do not exceed
10%, indicating similar, instead of vastly different, rates
of degradation of nC17 relative to Pr in the hydrocarbon
plume. Furthermore, our results and those of other in-
vestigators (Kaplan and Galperin, 1996; Schmidt, 1996,
1998; Kaplan et al., 1997; Smith et al., 2001; Wade, 2001;
Alimi, 2002; Stout et al., 2002; Wade, 2002; Kaplan,
2003) concur, in that environmental conditions more
commonly promote hydrocarbon degradation instead of
long-term preservation. The study of this relatively large
free product plume in the northwestern United States
supports the conclusion by Schmidt (1996, 1998) that
nC17/Pr ratios of the free product, like residual diesel
hydrocarbons in soil, decrease as a function of time.
Scenario 2: (nC17/Pr)o is not Known
If no information was available for a specific release
and (nC17/Pr)o could not be estimated using regional
refinery oil-supply data, a worst case scenario results,
in which, as discussed by Kaplan (2003), an experi-
enced investigator may only be able to constrain the
age of the release as being either older or younger than
about 10 yr, unless other site-specific information
were available to further refine the age. In the MDD
model, uncertainties in modeling the age of a release
increase as our confidence in (nC17/Pr)o ratios de-
crease. Specifically, differentiation of the relationship
between To and (nC17/Pr)o indicates that
dTo ¼ 9:76 dðnC17=PrÞo � 10 dðnC17=PrÞo
Hence, an uncertainty of ±0.1 in (nC17/Pr)o results in
an uncertainty of ±1 yr in the MDD model age. When
(nC17/Pr)o can only be assumed to lie in the range de-
fined by the statistical analyses of domestic crude oil
and refined product (nC17/Pr)o ratios presented in this
study, about 1.6 and 3.0 (2.3 ± 0.7), the uncertainty
in the MDD model ages becomes ±7 yr, which is the
worst case scenario.
Scenario 3: (nC17/Pr)o can be Calculated
If records are available, e.g., regional refinery oil sup-
ply data, that allow (nC17/Pr)o to be calculated, the
MDD model could be used and model ages adjusted
as previously discussed. Depending on the resulting
level of uncertainty in (nC17/Pr)o, concomitant model
age uncertainties would range from about ±2 to ±7 yr,
that is, an age uncertainty that lies between our best
and worst case scenarios.
Surface Releases
At sites involving surface hydrocarbon releases (i.e.,
aerobic conditions), measured nC17/Pr ratios have been
shown to produce erroneous C-L model ages (Smith
et al., 2001; Wade, 2001). However, this does not nec-
essarily compromise the use of nC17/Pr ratios as a
means of estimating the age of surface releases of mid-
dle distillates. Although degradation rates may be ac-
celerated, if the degradation rate is systematic, signif-
icant site-specific model ages can be determined. To
assess degradation rates and systematics of middle dis-
tillate and crude oil releases in surface environments,
data from the work of Smith et al. (2001), F. D. Hos-
tettler and K. A. Kvenvolden (1994, personal commu-
nication), and Schmidt (this study) are evaluated; these
data are plotted in Figure 3 (surface data points).
The case study employed by Smith et al. (2001) to
question the validity of the C-L model involved a sur-
face release of diesel fuel in Pennsylvania that occurred
in October 1993. Although the release was less than
3 yr old, ad hoc application of the C-L model to es-
timate the age of six free product samples collected
between 1993 and 1996 yielded erroneous ages that
Table 2. Downgradient (nC17/Pr) Ratios in a Diesel Fuel Plume
(nC17/Pr) at the Plume
Distance from Source Margin Interior
0 2.7; 3.0
250 m (820 ft) 2.5; 2.6 2.7
675 m (2214 ft) 2.0; 2.2 2.2
1045 m (3428 ft) 1.7; 1.7; 1.7
1350 m (4429 ft) 1.3; 1.3 1.3
186 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
ranged from about 1 to 18 yr, the majority exceeding the
known age of the release. A reevaluation of the data
presented by Smith et al. (2001) indicates that degrada-
tion of nC17 relative to pristane follows a well-defined
(R2 = 0.923) first-order kinetic reaction given by
ðnC17=PrÞt ¼ ðnC17=PrÞo e�lt
The parameters in this equation and their calculated
value as determined by the regression of the data from
Smith et al. (2001) are as follows: (nC17/Pr)t is the
(nC17/Pr) ratio of the degraded product t years after
the release; (nC17/Pr)o, the nC17/Pr ratio of the un-
degraded middle distillate at the time of the release
(T = 0) is 2.01; and l, the nC17/Pr decay constant, is 1.14.
The value of (nC17/Pr)o as calculated, 2.01, is in excel-
lent agreement with those observed, 2.0–2.2 (Smith
et al., 2001); and the decay constant, 1.14 per year, yields
a half-life, t1/2, for nC17/Pr of approximately 0.6 yr.
The erroneous C-L model ages led Smith et al. (2001)
to conclude that nC17/Pr ratios could not be used to
estimate ages of diesel releases, and that the C-L model
was scientifically invalid. However, although C-L model
ages are not correct, the degradation of nC17 relative
to pristane with time is systematic but has been ac-
celerated by the aerobic conditions.
Rapid degradation of organic compounds in aer-
obic environments, including compounds perceived
as being recalcitrant (e.g., polycyclic aromatic hydro-
carbons), has been observed in crude oil released into
Prince William Sound, Alaska, as a result of the 1989
Exxon Valdez disaster (Hostettler and Kvenvolden,
1994). Degradation of nC17 relative to Pr in Exxon
Valdez crude oil residues collected in 1990 follow a
first-order law that is indistinguishable from that de-
termined from the Smith et al. (2001) data (Figure 3).
Collectively, the Pennsylvania and Alaska data suggest
that the timescale of degradation of nC17 in diesel fuel
and crude oil released at the surface is approximately
5–6 yr; hence, age estimates using the original C-L or
MDD model would be erroneous.
Garden Experiment
To further investigate degradation rates of diesel fuel
in an aerobic environment, Schmidt performed a garden
experiment, in which a small volume of diesel fuel was
buried approximately 0.6 m (2 ft) below the ground
surface in nutrient-rich, sandy soil. Samples of the
diesel fuel were collected quarterly over a period of
about 2 yr and the (nC17/Pr) ratio of the degraded
diesel fuel present in the soil measured (Figure 3; Sur-
face OK). Although the (nC17/Pr)o of the diesel fuel in
the garden experiment differs from that observed in
Pennsylvania and Alaska diesel fuels (1.5 vs. 2.1, re-
spectively), the overall morphology of the (nC17/Pr)
time dependency curve is very similar to that observed
in the Pennsylvania–Alaska studies and again, nC17 in
the Oklahoma garden experiment will be totally de-
graded within 5–6 yr.
These results concur with those of previous stud-
ies (Bossert and Bartha, 1984; Alvarez and Vogel, 1995;
Kaplan and Galperin, 1996; Kaplan et al., 1997; Al-
varez et al., 1998; Morrison, 2000b; Smith et al., 2001),
which suggest that the rate of degradation of organic
compounds is accelerated in surface aerobic environ-
ments compared to those in the subsurface, where con-
ditions are less aerobic. Although age estimates of mid-
dle distillate and crude oil releases in highly aerobic
environments may be limited (�5–6 yr), the results
are neither meaningless nor do they necessarily inval-
idate age-dating models based on the degradation of
nC17 relative to Pr in petroleum hydrocarbons.
MEASURED (nC17/Pr)o RATIOS OF CRUDE OILSAND MIDDLE DISTILLATES
In the original Christensen and Larsen (1993) model,
11 fresh diesel samples were analyzed to evaluate var-
iations in (nC17/Pr)o ratios. Their results indicated an
average (nC17/Pr)o ratio of 1.98 with a 95% confidence
level standard error of the mean of 0.50 (hereafter
2ssem; calculated from their standard deviation of
0.83). As discussed later, an uncertainty of this mag-
nitude in (nC17/Pr)o alone doubles the uncertainty of
±2 yr in C-L model ages proposed by Christensen and
Larsen (1993; Hurst, 2003). Variations in (nC17Pr)o
and their potential impact on MDD model ages are
best addressed by first examining measured ranges in
(nC17/Pr)o ratios of crude oils and refined petroleum
hydrocarbons.
(nC17/Pr)o Ratios in Domestic and Foreign Crude Oils
Between 1965 and 2000, HRGC analyses have been
performed on thousands of foreign and domestic crude
oils, the vast majority being compiled by the petro-
leum industry in unpublished studies (e.g., Amoco
Crude Oil Analysis Program, 1965–1990; hereafter,
COAP); segments of these studies have been pub-
lished in the annals of special symposia (Schmidt,
1996, 1998). Results from the COAP investigation
Hurst and Schmidt 187
have recently been released (G. W. Schmidt, unpub-
lished data), and average (nC17/Pr)o ratios from this
extensive database are incorporated in this study. Our
analyses of the COAP database included 4541 domes-
tic and foreign crude oils, with the emphasis placed on
3549 crude oils from sources that provide about 85–
90% of the crude refined in the United States (Oil &Gas Journal, 2002; hereafter OGJ, 2002). Note that
average (nC17/Pr)o ratios of domestic and foreign crude
oils presented in Tables 3 and 4, respectively, are not
determined from a single producing zone in one field.
Instead, they were calculated using representative sam-
ples throughout active oil fields in regions either in the
United States or overseas. Hence, although crude oils
in a region may exhibit a range of (nC17/Pr)o ratios,
transport and storage of crude oil mixes the oil, reduc-
ing the observed range in (nC17/Pr)o ratios of the
resultant blended oil. Individual (nC17/Pr)o ratios for
about 3400 of the domestic plus foreign crude oils ana-
lyzed in our investigation are plotted in Figure 4.
Average (nC17/Pr)o ratios of 2737 domestic crude
oils, which account for about 90–95% of domestic
crude produced in the United States (OGJ, 2002) are
shown in Table 3. For reference, results have been
grouped using their Petroleum Administration De-
fense District (PADD), i.e., regional, designations. The
grand mean (nC17/Pr)o ratio for domestic crude oil
average (nC17/Pr)o ratios is 2.3 ± 0.7 (2ssem; Table 3);
the weighted mean, 1.9, which apportions each state’s
relative contribution to the total United States crude
oil production, is calculated as follows:
ðnC17=PrÞWTDo ¼
Xi½ðpi=PÞðnC17=PrÞo;i
where pi is the crude oil production in thousands of
barrels per day for a specific state (OGJ, 2002); P is
the total United States crude oil production (from
Table 3; P = 2737 � 103 bbl/day); and (nC17/Pr)o,i is
the average (nC17/Pr)o ratio for crude oil produced
in a specific state. Weighted means reduce the
impact of datum that could be considered as outliers,
such as North Dakota crude oil (minor producer with
an anomalously high crude oil (nC17/Pr)o ratio). For
comparison, the grand mean (nC17/Pr)o ratio and
related 2ssem for domestic crude oils, 2.3 ± 0.7, drop
to 2.0 ± 0.3, if the North Dakota crude oil datum point
is excluded (Table 3).
The grand mean (nC17/Pr)o ratio of foreign crude
oils produced by major importers, which account for ap-
proximately 75–80% of oil imported into the United
States, is identical to that of domestic crude oils, 2.3 ±
0.5 (2ssem; Table 4). Average (nC17/Pr)o ratios of Mex-
ican and the Middle Eastern crude oils tend to exceed
3.0, with Saudi Arabian crude oil exhibiting the high-
est ratio, 5.1; (nC17/Pr)o ratios of the 13 remaining for-
eign crude oils from Canada, South America, Africa,
and Eurasia range from 1.0 to 2.2, with seven exceed-
ing 1.8. Blending significant proportions of crude oil
from a major importer, such as Saudi Arabia or Middle
East in general, with domestic crude during refining in-
creases the (nC17/Pr)o ratio of both the blended crude
Table 3. Average (nC17/Pr)o Ratios of Domestic Crude Oils
State
Production
(thousands
bbl/day)* n* (nC17/Pr)
PADD 1: East CoastFlorida 10 7 2.2
PADD 2: Midwest*Kansas 93 123 2.4
Oklahoma 188 151 2.2
North Dakota 87 45 6.5
PADD 3: Gulf CoastLouisiana 1621 600 2.4
Mississippi 54 8 1.8
Texas 1364 744 1.9
New Mexico 186 55 1.5
PADD 4: Rocky MountainColorado 45 221 2.0
Montana 44 135 3.4
Utah 42 125 1.8
Wyoming 157 308 1.8
PADD 5: West CoastAlaska 963 96 1.9
California 799 119 1.0
Grand mean** 2737 2.3 ± 0.7
Grand mean
(North Dakota excluded)
2692 2.0 ± 0.3
Weighted mean** 2737 2.1
*Production figures given in thousands of barrels per day; n = number of crudeoils analyzed; PADD = Petroleum Administration Defense District.
**Grand mean calculated from the average (nC17/Pr)o ratios of the 13 UnitedStates crude oils tabulated; error is 2s standard error of the mean. Weightedmean is calculated using average 2001 production figures for each State (inthousands of barrels per day; Oil & Gas Journal, December 23, 2002).
188 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
Table 4. Average (nC17/Pr)o Ratios of Foreign Crude Oils
Location Country
Import Volume
(thousands bbl/day)* n* (nC17/Pr)
North America Canada 1856 287 1.9
Mexico 1440 37 3.4
South America Argentina no data 51 1.8
Venezuela (Organization
of Petroleum Exporting
Countries [OPEC])
1496 40 1.4
Middle East/Africa Algeria (OPEC) 273 33 2.2
Kuwait (OPEC) 248 3 4.4
Nigeria (OPEC) 768 24 1.0
Saudi Arabia (OPEC) 1622 17 5.1
Abu Dhabi no data 24 3.9
Iran no data 15 3.4
United Arab Republic no data 277 2.4
Gabon no data 49 1.3
Libya no data 33 1.9
EurAsia England 384 248 1.8
Norway 383 123 1.8
Azerbaijan no data 51 1.6
Indonesia no data 153 1.5
Pakistan no data 36 1.4
Russia no data 303 1.9
Imported crude oil grand mean** 1804 2.3 ± 0.5
*Imported crude oil volume given in thousands of barrels per day (bbl/day); n = number of crude oils analyzed. Average annual 2001 – 2002 import volumes areshown for nine crude oils for which data exist (Oil & Gas Journal, January 27, 2003).
**Grand mean calculated from the average (nC17/Pr)o ratios of the 19 imported crude oils tabulated; error is 2s standard error of the mean.
Figure 4. A histogram depict-ing 3401 (nC17/Pr)o ratios ofdomestic and foreign crudeoils from sources that provideabout 80–85% of the crudeoils refined in the United States.The data exhibit a normal dis-tribution in which the mean,1.9, is approximately equalto the mean 1.8. NAm = NorthAmerica; OPEC = Organizationof Petroleum Exporting Coun-tries; EurAsia = Europe andAsia.
Hurst and Schmidt 189
and refined product. Hence, drawing the conclusion that
products refined in California will have exceptionally
low (nC17/Pr)o ratios, reflecting those of California
crude, is not correct as discussed in the next section.
The worldwide crude oil data indicate the statisti-
cal averages, i.e., weighted and grand means, of domestic
and foreign crude oil (nC17/Pr)o ratios analyzed in this
study range from 1.9 to 2.3. The overall range determined
by incorporating the 95% confidence level standard er-
ror of the mean, ±2ssem, is about 1.6 to about 3.0. These
results are similar to (nC17/Pr)o ratios calculated from
the linear regression models of Christensen and Larsen
(1993), Schmidt (1996, 1998), Wade (2001), and this
study, which range from about 1.8 to 2.4 (Table 5,
bottom) when the 95% uncertainty is considered.
Crude Oil Blending: The California Example
If middle distillates in California were refined entirely
from California crude, the resultant (nC17/Pr)o ratios
of these products would be low, typically less than 1.5
and perhaps as low as 1.0. However, California refineries
blend crude oils from several sources that include Cal-
ifornia, Alaska, and foreign crude. The effect of blend-
ing crude and the impact on (nC17/Pr)o ratios of the
blended crude oil and subsequent refined products can
be demonstrated using data from the California Energy
Commission (CEC; http://www.energy.ca.gov).
The low (nC17/Pr)o ratios of California crude oils,
which average 1.0 but can range up to approximately
1.5 (Magoon and Isaacs, 1983; R. L. Ames, 2004,
personal communication), would result in significant
errors in MDD model ages if the model was applied
without consideration for variations in (nC17/Pr)o.
However, CEC statistics indicate that since about
1985, only about 45–50% of crude oil refined in
California originates from California oil fields. Since
1985, foreign sources of crude oil refined in California
have increased from about 5 to 25%, with the remaining
25–45% of the refined crude oil originating in Alaska.
Mexican and Middle Eastern crude oils with high
(nC17/Pr)o ratios are predominant (�70%), which
results in an average (nC17/Pr)o ratio of about 3.2 for
the foreign component of crude oil refined in Califor-
nia. Thus, the calculated average (nC17/Pr)o ratios of
crude oil refined in California, using CEC oil supply
source statistical data and Amoco COAP Project
results, has ranged from about 1.6 in the mid-1980s
to about 2.1 in 2001. These values exceed those of
unblended California crude oil yet are consistent with
those of 16 western United States middle distillates
analyzed in this study (1.8–2.0; Table 4). Concerns
Table 5. (nC17/Pr)o Ratios: Refined Products and Calculated
(Linear Regressions)
Location or State Product
Number
Analyzed (nC17/Pr)o
PADD 1: East CoastNortheast Diesel 29 2.1
PADD 2: MidwestIndiana Blend oil 1 2.0
Heavy naphtha 1 1.6
JP-8 1 1.9
Diesel 5 1.8
Jet A 2 2.0
Kerosene 1 2.1
Motor/lube oil 6 2.6
Fuel oil 2 2.1
Furnace oil 2 1.9
Oklahoma Kerosene 2 2.1
Diesel 2 2.3
Transmission oil 2 1.9
Michigan Diesel 1 1.4
Furnace oil 3 1.6
Minnesota Diesel 1 1.9
Tennessee Kerosene 1 2.0
PADD 3: Gulf CoastLouisiana Kerosene 1 1.4
Texas Jet A 1 1.7
Diesel 5 1.8
PADD 4: Rocky MountainColorado Diesel 1 2.5
Utah Diesel 3 1.8
PADD 5: West CoastArizona Jet A 1 4.0
Western United States Kerosene/jet A 2 1.8
Diesel 8 1.9
Fuel oil 6 2.0
United States petroleum product
grand mean
90 2.0 ± 0.2
(nC17/Pr)o calculated from linear regression models**
Christensen and Larsen (1993) 2.16 ± 0.22
Schmidt (1996) 2.16 ± 0.25
Wade (2001) 1.99 ± 0.13
MDD model 2.12 ± 0.12
*Grand mean calculated from the average (nC17/Pr)o ratios of the 26 petroleumproducts tabulated; error is 2s standard error of the mean (95% confidencelevel).
**For comparison, the (nC17/Pr)o ratio is calculated from each linear regressionmodel (see Table 1).
190 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases
regarding ad hoc applications of the C-L model and the
impact of uncertainties in (nC17/Pr)o on model ages are
well founded (Wade, 2001; Kaplan, 2003). However,
in some cases, uncertainties associated with (nC17/Pr)o
ratios may be reduced by integrating the data presented
in this study with region-specific oil supply statistics,
when available (Hurst, 2003).
(nC17/Pr)o Ratios of Domestic Middle Distillates andFuel Oils
Analyses of 90 middle distillate and fuel oil average
(nC17/Pr)o ratios analyzed between 1970 and 1990
are presented in Table 5 and plotted as individual da-
tum in Figure 5; PADD designations are again pro-
vided for reference. The average (nC17/Pr)o ratio of
these refined products, 2.0 (± 0.2; 2ssem), and the me-
dian, 1.9, indicate a normal distribution; the average
is also statistically indistinguishable from those ob-
served in crude oils analyzed in this study, indicating
that crude oil (nC17/Pr)o ratios are passed conserva-
tively to refined products (Schmidt, 1996, 1998; Mor-
rison, 2000b; Murphy and Morrison, 2002). Of the 90
refined products analyzed, approximately 85% of the
(nC17/Pr)o ratios lie between 1.6 and 2.6. The range
observed in (nC17/Pr)o ratios of refined products, i.e.,
1.3–4.9, when compared to that observed in crude oils,
0.1 to greater than 6.0 (Figures 4, 3, respectively) is
again an indication that blending of crude oils during
refining, as discussed previously, serves to reduce the
overall variation in (nC17/Pr)o ratios.
CONCLUSIONS
The degradation of nC17 relative to Pr, the MDD mod-
el, has the potential to constrain the age of crude oil,
diesel, jet A, and fuel oil releases, within certain limits,
into the environment. However, we reemphasize the
concerns stipulated in the scientific literature that due
diligence and care should be exercised prior to the ap-
plication of this, or any, model used to estimate ages of
petroleum hydrocarbon releases.
In this study, we have integrated both published
and new data from petroleum hydrocarbon releases
to evaluate the Christensen and Larsen (1993) model
at sites whose climatic conditions range from those
of Denmark, where the C-L model was originally de-
veloped, to sites in the southern United States and
the Caribbean with tropical climates. Our results in-
dicate that subsurface releases of crude oils, diesel, and
fuel oils do not degrade at a rate that is statistically
distinguishable from that of the original C-L model,
thereby allowing us to develop a unified model, the
MDD model, as a potential tool to estimate ages of such
releases.
Resultant age uncertainties associated with ap-
plications of the MDD model are dependent on local
environmental conditions and variations in (nC17/Pr)o
ratios. Under optimal conditions, MDD model age un-
certainties of ±1.5 yr may be achieved; as our confi-
dence in the value of (nC17/Pr)o ratios of the product
released and our knowledge of site conditions decrease,
MDD model age uncertainties increase to ±7 yr.
Figure 5. A histogram of(nC17/Pr)o ratios for 90 do-mestic refined products. Thedata define a normal distributionin which the mean � median,2.0 and 1.9, respectively.
Hurst and Schmidt 191
In aerobic environments, we observe that surface
releases of petroleum hydrocarbons weather rapidly,
with nC17 being totally degraded within about 6 yr.
Under such conditions, ad hoc application of the MDD
model is not appropriate given that erroneous ages
result (Smith et al., 2001); hence, the MDD model is
not applicable to surface releases. In the subsurface, if
conditions are more anaerobic, MDD model age un-
certainties can range from ±1.5 yr under optimal con-
ditions to ±5 yr in a worst case scenario, where little
information is available about the site and free product
(nC17/Pr)o ratios; in the latter case, it may only be
possible to conclude that the release is older or younger
than 10 yr.
Statistical evaluations of (nC17/Pr)o ratios analyzed
in approximately 4500 worldwide crude oils and 90 do-
mestic refined products (diesel, jet A, fuel oils) were
performed to evaluate the impact of variations in this
ratio on MDD model ages as well as to provide a re-
gional database for those who may wish to either apply
the MDD model to estimate the age of a release or
simply use the information in environmental forensic
investigations.
REFERENCES CITED
Alimi, H., 2002, Invited commentary on the Christensen andLarsen technique: Environmental Forensics, v. 3, p. 5.
Alvarez, P. J. J., and T. M. Vogel, 1995, Degradation of BTEX andtheir aerobic metabolites by indigenous microorganisms undernitrate reducing conditions: Water Science Technology, v. 31,p. 15–28.
Alvarez, P. J. J., R. C. Heathcot, and S. E. Powers, 1998, Cautionagainst interpreting gasoline release dates based on BTEX ratios:Groundwater Monitoring Review, v. 18, no. 4, p. 69–76.
Bossert, I., and R. Bartha, 1984, The fate of petroleum in soil eco-systems, in R. M. Atlas, ed., Petroleum microbiology: NewYork, New York, Macmillan Publishing Co., p. 435–447.
Brooks, C., S. R. Hart, and I. Wendt, 1972, Realistic use of two-errorregression treatments as applied to rubidium-strontium data:Reviews of Geophysics and Space Physics, v. 10, p. 551–577.
Christensen, L. B., and T. H. Larsen, 1993, Method for determiningthe age of diesel oil spills in soils: Groundwater MonitoringReview, Fall, p. 142–149.
Collins, W. D., 1925, Temperature of water available for industrialuse in the United States: U.S. Geological Survey Water SupplyPaper 520-F, 52 p.
Dyson, N., 1995, Modern data handling methods, in E. R. Adlard, ed.,Chromatography in the petroleum industry: Journal of Chro-matography Library Series, v. 56, p. 375–399.
Heath, R. C., 1984, Groundwater regions of the United States: U.S.Geological Survey Water Supply Paper, p. 22–42.
Hern, S. C., and S. M. Melancon, 1986, Vadose zone modelingof organic pollutants: Chelsea, Michigan, Lewis Publishers,295 p.
Hostettler, F. D., and K. A. Kvenvolden, 1994, Geochemical changes
in crude oil spilled from the Exxon Valdez supertanker intoPrince William Sound, Alaska: Organic Geochemistry, v. 21,p. 927–936.
Hurst, R. W., 2000, Applications of anthropogenic lead archaeostra-tigraphy, ALAS model, to hydrocarbon remediation: Environ-mental Forensics, v. 1, p. 11–23.
Hurst, R. W., 2002, Lead isotopes as age sensitive, genetic markersin hydrocarbons: 3. Leaded gasoline, 1923–1990: Journal of En-vironmental Geoscience, v. 9, p. 43–50.
Hurst, R. W., 2003, Invited commentary on Dr. Isaac Kaplan’s paper‘‘Age dating of environmental organic residues:’’ EnvironmentalForensics, v. 4, p. 145–152.
Kaplan, I. R., 2003, Age dating of environmental organic residues:Environmental Forensics, v. 4, p. 95–141.
Kaplan, I. R., and Y. Galperin, 1996, How to recognize a hydrocarbonfuel in the environment and estimate its age of release, in T. J.Bois and B. J. Luther, eds., Groundwater and soil contamina-tion: Technical preparation and litigation management: NewYork, New York, Wiley Law Publishers, p. 145–199.
Kaplan, I. R., Y. Galperin, S.-T. Lu, and R.-P. Lee, 1997, Forensicenvironmental geochemistry: Differentiation of fuel-types,their sources and release time: Organic Geochemistry, v. 27,p. 289–317.
Ludwig, K. R., 1983, Plotting and regression programs for isotopegeochemists: U.S Geological Survey Open-File Report 83-849,89 p.
Magoon, L. B., and C. M. Isaacs, 1983, Chemical characteristics ofsome crude oils from the Santa Maria basin, California, in C. M.Isaacs and R. E. Garrison, eds., Petroleum generation and oc-currence in the Miocene Monterey Formation, California:Pacific Section of the SEPM, p. 201–211.
Morrison, R. D., 2000a, Critical review of environmental forensictechniques: Part I: Environmental Forensics, v. 1, p. 157–173.
Morrison, R. D., 2000b, Critical review of environmental forensictechniques: Part II: Environmental Forensics, v. 1, p. 175–195.
Morrison, R. D., 2000c, Environmental forensics: Principles and ap-plications: New York, CRC Press, 351 p.
Murphy, B. L., and R. D. Morrison, 2002, Introduction to environ-mental forensics: New York, Academic Press, 560 p.
Schmidt, G. W., 1996, Forensic geochemistry for hydrocarbon re-leases: Proceedings of Petroleum Hydrocarbons and OrganicCompounds in Groundwater: Prevention, Detection and Re-mediation Conference, November 13–15, 1996, Houston, Texas:Westerville, Oklahoma, Groundwater Publishing Co., p. 549–572.
Schmidt, G. W., 1998, The effect of petroleum weathering on pat-tern recognition and dating: Proceedings of Environmental Fo-rensics: Determining Liability through Applied Science Con-ference, September 24–25, International Business Commission,Southborough, Massachusetts, p. 13.
Schmidt, G. W., and K. L. Marsi, 1961, Chemical analysis of groundwater resources of Rooks County, Kansas: Transactions of theKansas Academy of Science, v. 64, p. 49–62.
Smith, J. S., L. Eng, and D. A. Shepperd, 2001, Age-dating oil: IsChristensen and Larsen applicable: The Chemist, p. 9–13.
Stout, S., A. D. Uhler, K. J. McCarthy, and S. D. Emsbo-Mattingly,2002, Invited commentary on the Christensen and Larsen tech-nique: Environmental Forensics, v. 3, p. 9–11.
Van der Leeden, F., F. L. Troise, and D. K. Todd, 1990, The waterencyclopedia: Chelsea, Michigan, Lewis Publishers, 808 p.
Wade, M. J., 2001, Age-dating diesel fuel spills: Using the Europeanempirical time-based model in the United States: Environmen-tal Forensics, v. 2, p. 347–358.
Wade, M. J., 2002, Invited commentary on the Christensen andLarsen technique: Environmental Forensics, v. 3, p. 13.
192 Age Significance of nC17/Pr Ratios in Investigations of Product and Releases